Driving Turbulence in Simulations of Molecular Clouds

Star-forming giant molecular clouds are known to be turbulent but the origin of the turbulence is unclear. Modelling the formation of clouds and their continual buffeting by the surrounding ISM, which is likely to be the root cause, is extremely expensive, so many simulations model isolated clouds or small sections of ISM, and drive turbulence artificially within the whole volume of the simulation. This is clearly not realistic, because starless clouds have no internal energy source to drive turbulence internally.
The purpose of this project is to develop methods of driving turbulence only on the edges of a reference volume (to begin with, a cube representing a small region of the ISM, and later, an arbitrary gas distribution representing a single molecular cloud), and to examine the extent to which the turbulence permeates into the undriven reference volume. Simulations will initially be purely hydrodynamic, but the intention is to include magnetic fields as well, since they should aid in the coupling of the edges of the reference volume to its interior.